chore(gpu): pin ML deps in GPU Dockerfiles for stable CUDA/cu118 builds #6
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: ROCm Diagnostics (self-hosted) | ||
| true: | ||
| workflow_dispatch: {} | ||
| jobs: | ||
| diagnose: | ||
| name: Run ROCm diagnostics | ||
| runs-on: | ||
| - self-hosted | ||
| - linux | ||
| - rocm | ||
| steps: | ||
| - name: Checkout | ||
| uses: actions/checkout@v4 | ||
| - name: Runner and Docker info | ||
| shell: bash | ||
| run: 'uname -a | ||
| echo "Labels: ${{ runner.os }} ${{ runner.arch }}" | ||
| docker --version || true | ||
| docker compose version || docker-compose --version || true | ||
| ' | ||
| - name: Ensure script is executable | ||
| shell: bash | ||
| run: chmod +x scripts/diagnose-rocm.sh | ||
| - name: Run ROCm diagnostics | ||
| shell: bash | ||
| run: ./scripts/diagnose-rocm.sh --temp-container | ||
| - name: Upload diagnostics artifact | ||
| uses: actions/upload-artifact@v4 | ||
| with: | ||
| name: rocm-diagnostics-${{ github.run_id }} | ||
| path: 'logs/diagnostics/rocm-diagnose-*.log | ||
| ' | ||
| if-no-files-found: warn | ||
| retention-days: 14 | ||
